The 60-second answer
An NSF event — non-sufficient funds — is any moment your bank account did not have the funds to cover an attempted withdrawal. It shows up on the statement as a returned check, a returned ACH, or an overdraft fee. MCA underwriters count every one of them.
The count matters because a funder is about to start pulling money out of your account every business day for the next 6–18 months. Every NSF on your file is a data point telling them that the daily pull might fail. Too many and the funder will not write the deal — not because they think you are dishonest, but because the operational risk is too high.
The good news: NSF tolerance varies a lot by paper grade, and a meaningful number of NSFs can be excused with the right documentation.
The standard NSF tolerance by paper grade
Across the funders we track in 2026, the typical NSF tolerance bands look like this:
- A-paper: 0–3 NSFs across the most recent 4 months. Often 0 across the most recent 60 days. Premium pricing tier.
- B-paper: 4–7 NSFs across 4 months. Some funders allow up to 5 in the most recent month if balanced by strong revenue.
- C-paper: 8–15 NSFs across 4 months. Pricing premium of 6–12 factor points vs A-paper. Smaller offer sizes.
- D-paper / distressed: 15+ NSFs. Limited number of funders. Pricing premium of 15–25 factor points. Shorter terms. Often shorter advance amounts.
- Unfundable through mainstream channels: 25+ NSFs across 4 months, or a clear escalating pattern showing financial deterioration.
These bands are not universal. Each funder publishes their own credit policy and the specific thresholds shift over time. The general shape — A-paper is single digits, C-paper is double digits, D-paper is teens-plus — is durable.
How underwriters distinguish "manageable" NSFs from "systemic" NSFs
A funder does not just count NSFs; they look at the pattern.
- Distribution. 1 NSF in each of 4 months looks different from 4 NSFs all in the most recent month. The distributed pattern reads as normal operational noise. The clustered pattern reads as escalating stress.
- Trend. NSF count rising month over month is the worst signal — stronger negative than a flat-but-higher count. A high-but-declining NSF count can actually be a positive signal that the merchant is stabilizing.
- Amounts. Multiple small NSFs (under $500) on the same day from recurring vendors are often a single timing issue, not a real funds problem. A single large NSF on a tax payment is a real funds problem.
- Balance after the event. If the account recovered to a healthy balance within 24 hours of the NSF, the event was a timing miss. If the balance stayed negative for days, it was a real liquidity problem.
- Counterparty. An NSF on a vendor ACH is normal noise. An NSF on a prior MCA daily payment is catastrophic — it signals stacking failure and almost always kills the file.
The NSFs that can get excused
Not every NSF on your statement has to count against you. Funders will often exclude:
- Vendor double-debits. A single vendor accidentally debits twice on the same day, the second debit fails, the vendor refunds within 48 hours. Document with the vendor confirmation and the corrective deposit.
- Bank-side timing errors. The bank posts a withdrawal before posting a same-day deposit. Document with the deposit confirmation showing the same-day availability.
- R03 / R04 ACH returns. Codes that indicate "no account" or "invalid account" — typically means a vendor sent the ACH to the wrong account, not that the account had insufficient funds. The bank still codes the return as NSF in some statements. Document with the corrected ACH from the vendor.
- Pre-authorization holds that exceeded the actual charge. Common with hotel and rental car holds. The hold caused an apparent NSF that resolved when the actual smaller charge posted. Document with the booking confirmation.
The pattern for getting NSFs excused: short merchant note + the supporting document at intake. Asking the broker to plead later, after the stip request comes, almost never works as well as proactive documentation.
Worked example — same NSF count, different outcomes
Two merchants both have 6 NSFs across their 4-month file.
Merchant A: 2 NSFs in month one, 1 in month two, 2 in month three, 1 in month four. All under $300. All recovered within 24 hours. Pattern: stable. The underwriter scores this as normal operational noise on a B-paper file. Offer proceeds at standard B-paper pricing.
Merchant B: 0 NSFs in months one and two, 1 in month three, 5 in month four. Two of the month-four NSFs are on a prior MCA's daily payment. Pattern: escalating stress with stacking failure. The underwriter declines.
Same headline NSF count. Very different real-world meaning.
What lowers your NSF count over time
If your current file is over the NSF threshold for the funders you want, the most effective fixes:
- Move recurring vendor payments to ACH on dates that always have funds.Most NSFs are timing failures, not real funds shortages. Aligning the auto-debit dates with your highest-balance days eliminates them.
- Consolidate to one operating account. Multiple accounts increase the chance that a payment hits the wrong account on the wrong day.
- Open an overdraft line for genuine timing buffer. Most banks offer small overdraft credit lines on business checking. Used as a buffer, they eliminate the short-duration timing NSFs that drag down your file.
- Switch checks to cards or ACH where possible. Checks are the highest-NSF instrument because the funds-availability gap can be days.
- Wait two months and reapply. If the NSFs were truly a one-month cluster from a fixable cause, two clean subsequent months drop them out of the active 4-month window for most funders.
The signal NSFs send beyond the count
Even when an NSF count is technically within tolerance, the underlying pattern can shape how the funder prices the deal. A merchant with 2 NSFs that are both on prior debt servicing is priced very differently from a merchant with 2 NSFs that are both on a single misfired vendor debit. The pricing engine reads through to the cause where it can.
This is why proactive documentation matters even when you are well within tolerance. The 1-page note that explains your NSFs as timing-driven rather than liquidity-driven can shift pricing by 2–4 factor points.
The bottom line
NSFs are one of the most heavily-weighted single signals in MCA underwriting. The count matters; the pattern matters more; the explanation matters at the margin. Merchants who understand the thresholds, fix the fixable causes, and document the explainable events routinely fund through bands where merchants of the same NSF count would have been declined.
Frequently asked questions
- How many NSFs disqualify me from an MCA?
- A-paper funders typically cap at 3 NSFs across the most recent 4 months. B-paper funders go to 5–7. C-paper funders will go higher (10+) but with significant pricing premium. Above 15 in 4 months, the file is generally unfundable through mainstream channels.
- Does a returned ACH count the same as a returned check?
- Yes, in most funder systems. The bank statement codes them differently but the underwriting platform counts both as NSF events. The exception is ACH returns coded R03 (no account) or R04 (invalid account), which are sometimes excluded if proven to be vendor errors.
- Can a single NSF be 'forgiven' if I explain it?
- Sometimes. A one-time NSF from a known cause — a vendor double-debit, a check posted in the wrong order, a timing issue between deposit clearing and payroll — can be excluded if the merchant supplies a short note and the related transactions support the story.
- Does the funder count NSFs paid through overdraft protection?
- Yes. The negative balance event still appears in the statement and the OCR counts it. Overdraft-protected accounts mask the visual but not the data. Some funders weight them slightly less heavily.
- How long do NSF events stay on my file for future applications?
- MCA underwriting almost always uses the most recent four to six months only. NSFs from older statements are not re-pulled. However, your merchant history with a specific funder does include the prior file, so a renewal underwrite will reference the earlier NSF activity.